Abstract

Tungsten trioxide (WO3) was surface modified with Cu(II) nanoclusters and titanium dioxide (TiO2) nanopowders by using a simple impregnation method followed by a physical combining method. The obtained nanocomposites were studied by scanning electron microscope, X-ray photoelectron spectroscopy spectra, UV-visible light spectra, and photoluminescence, respectively. Although the photocatalytic activity of WO3 was negligible under visible light irradiation, the visible light photocatalytic activity of WO3 was drastically enhanced by surface modification of Cu(II) nanoclusters and TiO2 nanopowders. The enhanced photocatalytic activity is due to the efficient charge separation by TiO2 and Cu(II) nanoclusters functioning as cocatalysts on the surface. Thus, this simple strategy provides a facile route to prepare efficient visible-light-active photocatalysts for practical application.

1. Introduction

As one of the most important transition metal oxides, tungsten trioxide (WO3) has attracted considerable attention due to its promising physical and chemical properties [1, 2]. Considering its small band gap, stable physicochemical properties and resilience to photocorrosion effects, WO3 has been widely considered as a feasible candidate for visible-light photocatalysts [13]. However, several fundamental issues have to be addressed before they are economically available for large scale industrial applications. For example, pure WO3 is usually not efficient photocatalysts because of the high electron-hole recombination rate and the difficulty in the reduction of oxygen, due to the negative position of its conduction band (CB) [4]. Thus, many efforts have been made to improve the activity of WO3, such as morphology control, doping, nanostructure construction, and surface modification [4]. One of the most promising ways to accomplish this goal is to design heterogeneous catalysts [5]. So far, various heterogeneous WO3 based heterogeneous structures, such as WO3/SiO2, WO3/TiO2, WO3/NiO, and Pt/TiO2-WO3, have been designed toward good catalytic performance [512].

Titanium dioxide (TiO2) has attracted much attention as a suitable semiconductor to construct heterogeneous structures with WO3, due to its low cost, nontoxicity, and suitable band structure [5, 1316]. The valence band (VB) and CB potentials of TiO2 are more cathodic than those of WO3 [16]. The coupling of TiO2 and WO3 can lead to photogenerated electron and hole transfer from one semiconductor particle to another; those are electrons transfer from the CB of TiO2 down to the CB of WO3 and holes transfer from the VB of WO3 to that of TiO2 [17]. This process suppresses the recombination of photogenerated carriers and leads to improved photocatalytic efficiency of the system [17]. To further increase the photocatalytic performance of the system, metal or metal oxide particles, such as Au, Pt, and CuO, were introduced into the system to promote the reduction reaction of electrons with oxygen molecules, leading to efficient consumption of electrons [1012]. For example, copper ions modified WO3/TiO2 nanocomposites, prepared by Hosogi and Kuroda, exhibited efficient photocatalytic activities [12]. However, there are still many problems in the practical application of these reported catalysts, including the complexity of the preparation procedures, the requirement for expensive raw materials, and the difficulties for large scale production [512]. For instance, sol-gel method and urea as raw materials were needed in the preparation of copper ions modified WO3/TiO2 nanocomposites [12]. Thus, efforts aimed at improving the photocatalytic performance and the preparation process of WO3 are still needed.

In the present work, we reported efficient Cu(II) nanoclusters modified WO3/TiO2 nanocomposites (Cu(II)-WO3/TiO2) through a facile preparation process. In this process, Cu(II) nanoclusters were deposited on WO3 using a simple impregnation method and TiO2 nanopowders were introduced into the Cu(II) nanoclusters modified WO3 (Cu(II)-WO3) by a physical combination method. The introduced Cu(II) nanoclusters and TiO2 nanopowders functioned as cocatalysts on the surface of WO3. Thus, the obtained Cu(II)-WO3/TiO2 products exhibited an enhanced visible light photocatalytic activity.

2. Experimental

2.1. Materials

Commercial tungsten (VI) oxide (Sigma-Aldrich; for monoclinic WO3, particle size is ~100 nm) was used as the initial WO3. CuCl2·2H2O (Sigma-Aldrich) was used as the source of Cu(II) nanoclusters. Degussa (Evonik) P25 TiO2 nanopowders (particle size ~25 nm) was used as the raw material of TiO2. All of these commercial materials were used as received, without further purification. Distilled water was applied in the experimental process.

2.2. Preparation of the Composites

Cu(II) nanoclusters were grafted on the surface of WO3 by using an impregnation method, as reported previously [18, 19]. CuCl2·2H2O was used as the Cu(II) nanoclusters source to prepare Cu(II)-WO3. 1 g WO3 powder with 0.1% weight fraction of Cu to WO3 was dispersed in 10 mL distilled water. 0.1% weight fraction of Cu to WO3 has demonstrated the optimized amount of Cu(II) nanoclusters for Cu(II)-WO3 systems [18]. The suspension was heated at 90°C and stirred for 1 h in a vial reactor to hydrolyze the CuCl2 source and generate Cu(II) nanoclusters on the surface of WO3. Then, the suspension was filtered twice with a membrane filter (0.025 μm, Millipore) and washed with sufficient amounts of distilled water. The resulting residue was dried at 110°C for 24 h and subsequently grounded into fine powder using an agate mortar and pestle.

The mixing of Cu(II)-WO3 with TiO2 was performed using a physical mixing method. Typically, 1 g Cu(II)-WO3 powder and 1% weight ratio TiO2 were mixed in an agate mortar and grounded into fine powder using a pestle for 1 h.

2.3. Sample Characterizations

Scanning electron microscope (SEM) images were taken using a field-emission SEM (FE-SEM, Hitachi S-4800). Photoluminescence PL spectra were obtained by using a Hitachi F-4500 fluorophotometer with an excited wavelength of λ = 325 nm at room temperature. UV-visible reflectance spectra were obtained by the diffuse reflection method using a spectrometer (UV-2550, Shimadzu). Surface compositions were studied by X-ray photoelectron spectroscopy (XPS; model 5600, Perkin-Elmer). The binding energy data were calibrated with reference to the C 1s signal at 284.5 eV.

2.4. Catalytic Activity Testing

Photocatalytic activity of the WO3 samples was evaluated in terms of the decolorization of methylene blue (MB) dye under visible irradiation. 20 mg sample was dispersed into 100 mL of 10 mg/L MB solution and stirred in the dark for 1 h to reach a complete adsorption-desorption equilibrium. Then the solution was irradiated with ~20 mW/cm2 visible light (>420 nm, with a light filter L42 (Asahi Techno-Glass)) under continuous stirring. With a given irradiation time interval, some specimens (5 mL) were taken from the dispersion and were centrifuged (4000 rpm). The clear upper solution was subjected to UV-Vis spectrophotometer (UV-2550, Shimadzu). The concentration of MB was determined from the absorbance at the wavelength of 665 nm.

3. Results

Figure 1 shows the SEM images of the obtained samples. It can be seen that the bare WO3 samples contained many particles. These particles have a clear surface and a size of ~100 nm. After Cu(II) nanoclusters grafting, the particle morphology still remained, indicating the grafting of Cu(II) nanoclusters did not affect its morphology (Figure 1(b)). After the modification with TiO2 nanopowders, some small particles could be observed in TiO2 mixed Cu(II)-WO3 (Cu(II)-WO3/TiO2) samples (Figure 1(c)). These small particles have a size around several tens of nanometers, coinciding with the size of TiO2 nanopowders. In the paper, except specially noted, the weight ratios of Cu and TiO2 to WO3 were set to 0.1% and 1%, respectively,

In order to determine the surface composition and chemical states of the surface elements, XPS spectra were recorded, as shown in Figure 2. In the W 4f and O 1s core-level spectra of the samples (Figures 2(a) and 2(b)), no obvious differences could be seen in the chemical states of elements W and O, demonstrating that neither the surface grafting of Cu(II) nanoclusters nor physical mixing of TiO2 powders affected the bonding structure between tungsten and oxygen. In the Cu 2p core-level spectra (Figure 2(c)), Cu signals were clearly observed in Cu(II)-grafted samples, such as Cu(II)-WO3 and Cu(II)-WO3/TiO2, confirming that Cu(II) was successfully grafted on the surface of WO3, while, in the Ti 2p core-level spectra (Figure 2(d)), Ti signal was only observed in Cu(II)-WO3/TiO2 composites, indicating the TiO2 was well mixed with WO3 powders.

Figure 3 shows the UV-Vis of the samples. It clearly shows that WO3 has a good visible light absorption property, indicating it is a potential visible light photocatalyst. The absorption edge of WO3 is located at ~460 nm, which corresponds to the interband transition of WO3 [18]. This interband absorption indicates a band gap of ~2.7 eV, which coincides with the reported values of 2.7 eV for WO3 [20, 21]. After Cu(II) nanoclusters grafting, an additional light absorption at the range of ~700–800 nm was clearly superimposed on the light absorption of WO3, as shown in the inset of Figure 3. This additional light absorption can be attributed to the d-d transition of Cu(II) [18]. After further modification with TiO2 powders, the interband transition of WO3 was not changed, due to the small amount of TiO2 powders and their large band gap [1315]. Notably, the additional visible light absorption caused by the d-d transition of Cu(II) can still be observed in the mixed nanocomposites, proving the existence of Cu(II) nanoclusters (inset of Figure 3).

Figure 4 represents the variation of MB concentration by photocatalytic reaction with the samples under visible light (>420 nm) irradiation. Typical evolution of MB concentration during photocatalytic reaction on Cu(II)-WO3/TiO2 is presented in Figure 4(a). Under light irradiation, the characteristic MB absorption peak decreased sharply and almost no color was observed after 90 minutes of irradiation, indicating that MB was completely degraded by Cu(II)-WO3/TiO2. Comparative studies among bare WO3, Cu(II)-WO3, and Cu(II)-WO3/TiO2 show that bare WO3 has a negligible activity under visible light irradiation (Figure 4(b)). The grafting of Cu(II) nanoclusters to the surface switched its photocatalytic activity. It can be seen that MB dye was almost degraded by Cu(II)-WO3 with 2 h of visible light irradiation. Interestingly, after further modification with TiO2 nanopowders, Cu(II)-WO3/TiO2 exhibited an enhanced photocatalytic activity compared with that of Cu(II)-WO3. MB dye was completely degraded by Cu(II)-WO3/TiO2 nanocomposites with 1.5 h of visible light irradiation, revealing the high photocatalytic activity of the Cu(II)-WO3/TiO2 nanocomposites. Figure 4(c) shows the pseudo-first-order kinetic rate for the photochemical degradation of MB by Cu(II)-TiO2 samples. The pseudo-first-order kinetic rate was calculated according to the equation of , where is the normalized MB concentration, is the reaction time, and is the pseudo-first-rate constant. It can been seen that the Cu(II)-WO3/TiO2 samples presented the highest reaction rate. The reaction rate was sharply decreased when bare WO3 was used. The result was consistent with the MB decomposition curves in Figure 4(b). Figure 4(d) shows the cycling measurements of MB decomposition over Cu(II)-WO3/TiO2. Similar values were obtained after 5-cycle measurements, suggesting a good stability for the photocatalytic application of Cu(II)-WO3/TiO2.

We also investigated the influences of experimental parameters on the photocatalytic performances of Cu(II)-WO3/TiO2 samples. Figure 5 shows the photocatalytic performances of Cu(II)-WO3/TiO2 samples with different ratios of TiO2. It can be seen that the activity of the samples was increased with the ratio of TiO2 to WO3 in the beginning. After the highest activity was achieved at the ratio of 1%, the photocatalytic activity was decreased again with the increase of ratio. These results revealed that the Cu(II)-WO3/TiO2 samples obtained with 1% TiO2 have the optimum amount of TiO2 for hole separation and reaction. It has been reported that the amount of TiO2 to mix with WO3 was important for the photocatalytic reaction [22, 23]. Thus, the Cu(II)-WO3/TiO2 samples with a TiO2 ratio of 1% exhibited the highest photocatalytic performance.

4. Discussions

Figure 6 shows the energy levels of TiO2 and WO3 [18]. TiO2, as one of the most efficient photocatalysts, has a high potential CB and a deep VB. Thus, electrons in its CB have sufficient reduction power for oxygen reaction with single electron and holes in its VB have large oxidation power for organic compounds decomposition, respectively. Consequently, TiO2 has a very high efficiency for photocatalytic reactions under UV light irradiation. However, TiO2 can only be activated under UV light irradiation, owing to its large band gap. WO3 is sensitive to visible light because of its proper band gap, 2.7 eV [20, 21]. Notably, both the CB and VB positions of WO3 are more positive than those of TiO2. As a result, photogenerated electrons can be transferred from the CB of TiO2 to the CB of WO3 and photogenerated holes can be transferred from the VB of WO3 to that of TiO2 [17]. Moreover, if photons do not have enough energy to excite TiO2 but have enough energy to excite WO3, hole in the VB of WO3 is still possibly transferred to the VB of TiO2 [24]. This process suppresses the recombination of photogenerated carriers and indicates that TiO2 can act as hole cocatalyst [12, 17, 24]. On the other hand, the CB potential of WO3 is lower than the potential for reduction reaction of oxygen molecules, leading to the insufficient consumption of electrons in CB. When Cu(II) nanoclusters were modified on the surface of WO3, the photogenerated electrons in the CB of WO3 can be transferred to the Cu(II) nanoclusters. The transferred electrons can be consumed by multielectron reduction reactions with oxygen molecules in the Cu(II) nanoclusters [18, 19]. In other words, Cu(II) nanoclusters function as efficient electron cocatalysts [12, 18, 19]. Consequently, the activity of Cu(II)-WO3 can be further enhanced by combining with TiO2.

Figure 7 shows the PL spectra of bare WO3, Cu(II)-WO3, and Cu(II)-WO3/TiO2, respectively. The main emission peak for WO3 is centered at about 460 nm, which is approximately equal to the band gap energy of WO3 [20, 25]. It worth noting that bare WO3 exhibited the highest PL intensity among these samples, indicating the highest recombination rate of electrons and holes [26]. After the Cu(II) nanoclusters were grafted, the intensity of the PL emission decreases, which can be attributed to the decrease of the efficient electron trapping and consumption on Cu(II) nanoclusters [18, 19]. The emission intensity of the Cu(II)-WO3/TiO2 was lower than that of bare WO3 and Cu(II)-WO3, which indicated that the recombination rate of photogenerated charge carriers was the lowest in the Cu(II)-WO3/TiO2. The PL results confirmed the importance of the modification of Cu(II) nanoclusters and TiO2 nanopowders for hindering the recombination of electrons and holes. Thus, efficient visible light photocatalytic activity can be achieved in Cu(II) and TiO2 modified WO3.

5. Conclusions

Efficient WO3 photocatalysts were prepared by being simply surface modified with Cu(II) nanoclusters and TiO2 nanopowders. In this prepared system, Cu(II) nanoclusters and TiO2 nanopowders were deposited on the surface of WO3 using a simple impregnation method and a physical combination method, respectively. Cu(II) nanoclusters and TiO2 nanopowders functioned as efficient cocatalysts on the surface of WO3, which acted as photocatalyst. Thus, efficient charge separations and reactions can be achieved in this Cu(II)-WO3/TiO2 system, resulting in efficient visible light photocatalytic reaction for organic compounds decomposition. The simple strategy opens an avenue for designing efficient visible-light-active photocatalysts for practical application.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgments

This work was supported by the Scientific Research Fund of Hunan Provincial Education Department under Grant no. 09A083 and the Science and Technology Fund of Hunan Provincial Science and Technology Department under Grant no. 2012FJ4137.